Ecdysone signaling at metamorphosis triggers apoptosis of Drosophilaabdominal muscles

Jonathan Zirin a,n, Daojun Cheng a,c, Nagaraju Dhanyasi e,f, Julio Cho a, Jean-Maurice Dura d,Krishnaswamy VijayRaghavan e, Norbert Perrimon a,b,nQ1

a Department of Genetics, Harvard Medical School, USAb Howard Hughes Medical Institute, 77 Avenue Louis Pasteur, Boston, MA 02115, USAc State Key Laboratory of Silkworm Genome Biology, Southwest University, Chongqing, Chinad Institute of Human Genetics, UPR 1142, CNRS. 141, rue de la Cardonille, 34396 Montpellier, Francee National Centre for Biological Sciences (NCBS)-TIFR, Bangalore, Indiaf Weizmann Institute of Science, Rehovot, Israel

a r t i c l e i n f o

Article history:Received 3 May 2013Received in revised form12 August 2013Accepted 19 August 2013

KeywordsQ3 :MetamorphosisMuscleDrosophilaHistolysisApoptosisAutophagyEcdysoneFtz-f1

a b s t r a c t

One of the most dramatic examples of programmed cell death occurs during Drosophila metamorphosis,when most of the larval tissues are destroyed in a process termed histolysis. Much of our understandingof this process comes from analyses of salivary gland and midgut cell death. In contrast, relatively little isknown about the degradation of the larval musculature. Here, we analyze the programmed destructionof the abdominal dorsal exterior oblique muscle (DEOM) which occurs during the first 24 h ofmetamorphosis. We find that ecdysone signaling through Ecdysone receptor isoform B1 is required cellautonomously for the muscle death. Furthermore, we show that the orphan nuclear receptor FTZ-F1,opposed by another nuclear receptor, HR39, plays a critical role in the timing of DEOM histolysis. Finally,we show that unlike the histolysis of salivary gland and midgut, abdominal muscle death occurs byapoptosis, and does not require autophagy. Thus, there is no set rule as to the role of autophagy andapoptosis during Drosophila histolysis.

& 2013 Published by Elsevier Inc.

Introduction

Programmed cell death (PCD) describes the process by which acell, through a regulated sequence of events, kills itself. Thisprocess is deployed in both plants and animals as a response tocell stress and as a means to eliminate extraneous cells duringdevelopment (Conradt, 2009; Lockshin and Zakeri, 2007). PCD canrefer to apoptosis, autophagy or programmed necrosis. Apoptosis,referred to as type I PCD, is characterized by caspase activation,cell shrinkage, membrane blebbing, nuclear condensation andDNA fragmentation, and finally loss of adhesion to extracellularmatrix and phagocytic engulfment (Kerr et al., 1972). Autophagy,referred to as type II PCD, involves bulk degradation of organellesand long-lived proteins. Upon autophagy induction, a doublemembrane vesicle envelops the cytoplasm, then fuses with thelysosome, where its inner membrane and contents are degraded

by hydrolases (Yang and Klionsky, 2009). Autophagy is not strictlya cell death pathway, as it facilitates the recycling of essentialnutrients and energy during periods of starvation, and protectsthe cell from injuries caused by stress and infection. Finally,programmed necrosis, also known as type III PCD is a non-lysosomal type of cell death that has characteristics of necrosis,but is rarely observed (Clarke, 1990).

One of the most dramatic examples of PCD occurs duringDrosophila metamorphosis. After hatching, the Drosophila larvaspends the next 3 to 4 days feeding, growing and developing theimaginal tissues and other primodia which will ultimately formmuch of the adult body. During metamorphosis, most of the larvaltissues are destroyed through PCD in a process termed histolysis(Baehrecke, 1996). The transitions between developmental stagesin Drosophila are controlled by pulses of the steroid hormone20-hydroxyecdysone (ecdysone) (Riddiford et al., 2000; Thummel,2001). A high titer pulse of the hormone triggers the end of thelarval stage and the transition to the puparium phase, and asubsequent pulse 12 h later triggers head eversion marking thebeginning of the pupal phase (Thummel, 1996). Pupation lastsapproximately 3.5 days, after which the adult fly emerges from thepupal case.

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

6768697071727374757677787980818283848586878889909192

Contents lists available at ScienceDirect

journal homepage: www.elsevier.com/locate/developmentalbiology

Developmental Biology

0012-1606/$ - see front matter & 2013 Published by Elsevier Inc.http://dx.doi.org/10.1016/j.ydbio.2013.08.029

n CorrespondingQ2 authors at: Department of Genetics, Harvard Medical School,USA. Fax: þ617 432 7688.

E-mail addresses: jzirin@genetics.med.harvard.edu,jonathanzirin@yahoo.com (J. Zirin),perrimon@receptor.med.harvard.edu (N. Perrimon).

Please cite this article as: Zirin, J., et al., Ecdysone signaling at metamorphosis triggers apoptosis of Drosophila abdominal muscles. Dev.Biol. (2013), http://dx.doi.org/10.1016/j.ydbio.2013.08.029i

Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎

Our understanding of the signaling pathway leading to histolysiscomes primarily from analyses of the salivary glands, which aredestroyed in response to the ecdysone peak at the prepupal/pupaltransition, respectively (Lee and Baehrecke, 2001; Lee et al., 2003; Leeet al., 2002b). Ecdysone binds to a heterodimeric receptor complexconsisting of ecdysone receptor (EcR) and ultraspiracle (USP) (Koelleet al., 1991; Thomas et al., 1993; Yao et al., 1992). This complex entersthe nucleus and activates the transcription of the so-called early genes,including Broad-Complex (BR-C), E74A, E75, and E93 (Baehrecke andThummel, 1995; Burtis et al., 1990; DiBello et al., 1991; Segraves andHogness, 1990). The early genes, in turn activate transcription of alarger set of late genes/secondary response genes, which are thoughtto dictate the specific developmental output of the ecdysone signalingcascade (Thummel, 1995, 1996).

Another important player in the timing of salivary gland celldeath is the orphan nuclear hormone receptor gene ftz-f1 (βftz-f1),which is transcribed midway through prepupal development,when the ecdysone titer is relatively low (Lavorgna et al., 1993).In salivary glands, FTZ-F1 functions together with EcR to activatetranscription of the early genes and subsequent cell death path-way in response to the ecdysone pulse at the prepupal/pupaltransition (Broadus et al., 1999; Woodard et al., 1994). In additionto its role in salivary gland death, FTZ-F1 was recently shown toplay a critical role during the ecdysone-dependent remodeling offat body cells (Bond et al., 2011) mushroom body neurons, andmotor neuronsduring metamorphosis (Boulanger et al., 2011,2012; Redt-Clouet et al., 2012; Sullivan and Thummel, 2003).

Much of the evidence for autophagic cell death comes fromstudies of salivary gland cell death during metamorphosis, whichrequires autophagy, although it remains unclear precisely howautophagy promotes death (Baehrecke, 2005; Berry andBaehrecke, 2007; Das et al., 2012). Additional evidence comesfrom studies of the midgut, which is destroyed in response tothe ecdysone peak at the end of larval development (Leeet al., 2002a). Like salivary gland death, midgut histolysis alsorequires a functioning autophagy pathway (Denton et al., 2009,2010). Less well studied than either salivary gland or midguthistolysis is the destruction of the larval muscles, which occursfirst in the thoracic segments at the onset of puparium formation,then about 8–10 h later in the abdominal muscles (Bodenstein;Robertson, 1936).

Like the salivary gland and midgut, Drosophila produces distinctlarval and adult sets of muscles. In the first 24 h after pupariumformation (APF) most of the larval muscles have been completelydegraded (Bate et al., 1991; Currie and Bate, 1991). A small set ofmuscles persist throughout metamorphosis, serving as templates forthe morphogenesis of adult muscles, then die in the first two daysafter the adult fly ecloses (Dutta et al., 2002; Fernandes et al., 1991;Kimura and Truman, 1990; Roy and VijayRaghavan, 1998). Amongthese muscles are the dorsal internal oblique muscles (DIOMs), whichlie just underneath the dorsal external oblique muscles (DEOMs) ineach hemisegment of the abdomen (Currie and Bate, 1991; Wasseret al., 2007). These latter muscles are one of the few cases where theprocess of cell death has been examined in any detail during the firstday of metamorphosis, when the vast majority of all larval muscles aredestroyed (Wasser et al., 2007). Here, we extend these studies,answering major questions about the process of muscle histolysis,including the role of the ecdysone pathway and the roles thatapoptosis and autophagy play in carrying out PCD. We use the DEOMsas a model to examine these issues in more detail, and show thatalthough both autophagy and apoptosis occurs during DEOM histo-lysis, only the latter is required for muscle degradation. We show thatdestruction of these muscles is dependent on ecdysone signalingthrough the EcR-B1 isoform. Further, we show that FTZ-F1, opposed byanother nuclear receptor, HR39, plays a critical role in the timing ofDEOM histolysis.

Materials and methods

Fly stocks

Fly stocks used were: yw; Dmef2-Gal4 (Ranganayakulu et al., 1996);Dmef2-Gal4 was recombined with UAS-GFP-Atg8a (Juhasz et al., 2008);Hr3903508 (Hr39-lacZ) is a lethal lacZ enhancer trap insertion (Allenand Spradling, 2008); Hr39neo8 (Allen and Spradling, 2008); UAS-Hr39(Boulanger et al., 2011); UAS-ftz-f1 (Boulanger et al., 2011); MhcweeP26

(Clyne et al., 2003); ftz-f1 protein trap (YB0007), from Flytrap (flytrap.med.yale.edu) (Buszczak et al., 2007). The following RNAi lines wereobtained from the DRSC/TriP DRSC/TRiP at Harvard Medical School(http://www.flyrnai.org/TRiP-HOME.html): white (HMS0004); Atg1(JF02273); Atg5 (HMS01244); Atg18 (JF02898), and ftz-f1 (JF02738and HMS00019). Additional fly lines were obtained from the Bloo-mington stock center: UAS-p35 (BL5073); UAS-EcR-B1DN (BL6872);UAS-EcR-B1 RNAi (BL9329); UAS-EcR RNAi104 (targets all isoforms ofEcR) (BL9327). Unless otherwise noted, flies were reared at 25 1C.

Immunostaining and antibodies

For whole-mount immunostaining of fly tissues, 3rd instarlarval, prepupal, or pupal body wall muscles were dissectedaccording to (Budnik et al., 2006) and fixed for 15 min in PBSwith 4% formaldehyde. After washing in PBT (1X PBSþ0.1% TritonX-100), samples were incubated overnight with the followingantibodies (in PBT): rabbit anti-cleaved caspase 3, 1:300 (CellSignaling Technology); mouse anti-EcR-A (15G1a), 1:50 (Develop-mental Studies Hybridoma Bank, DSHB); mouse anti-EcR-B1(AD4.4), 1:50 (DSHB); mouse anti-EcR-Common (DDA2.7), 1:50(DSHB); rabbit anti-beta-galactosidase (anti β-Gal), 1:2000 (Cap-pel).. After incubation with primary antibodies, the samples werewashed in PBT and incubated with Alexa-conjugated secondaryantibodies (Molecular Probes, 1:1000) and Alexa 488, 594, or 633-conjugated phalloidin (1:1000) to visualize F-actin (MolecularProbes, 1:1000). Nuclei were visualized by DAPI staining (1 μg/ml). Samples were washed in PBT and mounted in 1:1 glycerol/PBSand images were acquired with a Leica SP2 laser scanning confocalmicroscope. For pupal staging, white prepupae were individuallyselected at 0 h APF and age was determined relative to this point.All animals were maintained at 25 1C.

Feeding protocol, and chloroquine treatment

Unless otherwise noted, all fly crosses and larvae were main-tained in vials containing ‘standard’ food composed of 16.5 g/Lyeast, 9.5 g/L soy flour, 71 g/L cornmeal, 5.5 g/L agar, 5.5 g/L malt,7.5% corn sirup, 0.4% propionic acid, and 1.3% Tegosept. Forstarvation induced autophagy experiments 3rd instar larvae wereindividually selected and no more than 20 per experiment weretransferred to ‘low nutrient food’ composed of 0.3X standard food.Likewise, for the rich food diet, no more than 20 larvae weretransferred to ‘high nutrient food’ composed of standard foodsupplemented with 100 g/L sucrose and 50 g/L yeast. For eachgenotype, at least 4 larvae from at least two independent vialswere analyzed. For drug treatments, chloroquine diphosphate salt(Sigma) was added to the food at 2.5 mg/ml.

TUNEL

TUNEL labeling was performed according to the manufacturerprotocol (In Situ Cell Death Detection Kit, Roche). Briefly, prepupaewere dissected in PBS, then fixed in 4% formaldehyde in PBS for15 min. After being washed and permeabilized in PBT, the muscleswere incubated for 45 min at 37 1C in the TUNEL reaction. Samples

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132

J. Zirin et al. / Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎2

Please cite this article as: Zirin, J., et al., Ecdysone signaling at metamorphosis triggers apoptosis of Drosophila abdominal muscles. Dev.Biol. (2013), http://dx.doi.org/10.1016/j.ydbio.2013.08.029i

were then rinsed 3X in PBT and 3X in PBS, mounted, and analyzedby confocal microscope.

Transmission electron microscopy

White pre-pupae were marked and aged to the required devel-opmental stage. Pupal muscles were dissected in Schneider's insectmedium and fixed using paraformaldehyde and glutaraldehyde(2.5% each) in 0.1 M sodium cacodylate buffer (pH 7.4), followedby three washes in buffer alone. The samples were secondarilystained with 1%OsO4 for 2 h, washed and incubated with 2% uranylacetate in 50% alcohol. The samples were then taken through analcohol dehydration series (50, 75, 94 and 100%), infiltrated usingpropylene oxide and resin, and embedded and polymerized usingEpon 812 (EMS) for 48 h. Ultra-thin sectioning (70–90 nm) wasperformed using a Leica Ultra microtome UCT. Sections were stainedwith 1% aqueous uranyl acetate and lead citrate solutions, mountedon grids and imaged using a Tecnai T12 Spirit transmission electronmicroscope (FEI) equipped with an Eagle CCD camera (FEI). TIAsoftware was used for image acquisition.

Results

Time course of abdominal muscle degeneration

To fluorescently mark the muscles and observe their morphologystarting at pupariation, we used a transgenic fly line (weeP26)expressing a GFP-tagged Myosin heavy chain protein (Fig. 1,Fig. S1). Up to 4–6 h APF, neither the DIOMs nor the DEOMs showany signs of deterioration (Fig. 1A, B). At 8 h APF, however, weobserved the first signs of degradation in the DEOMs (Fig. 1C), which,unlike the DIOMs, begin to round up with disorganized sarcomerestructure (marked by GFP-Mhc). By 12 h APF, following head eversionand the transition from prepupa to pupa, the DEOMs are almostcompletely degraded, with most of the tissue fragmented anddispersed (Fig. 1D). At this stage DIOM morphology and GFP-Mhcpattern remain largely intact, although the muscles are thinner andmore compact compared to earlier timepoints.

To obtain a more detailed view of the changes in musclestructure during histolysis of the DEOMs compared to the DIOMs,we analyzed the muscles by transmission electron microscopy(EM). At 0 h APF, both the DIOMs and DEOMs have well-orderedsarcomere structures, with clearly defined Z-disc, A-band and

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132

Fig. 1. Time course of degenerating and persistent muscles during metamorphosis. (A–D) Abdominal muscles from the weeP26 transgenic line, expressing a GFP-taggedMyosin heavy chain protein (green). The dorsal external oblique muscle (DEOM) is marked with a red outline. The dorsal internal oblique muscle (DIOM) is marked with blueoutline. (A) At 0 h and (B) 4–6 h after puparium formation (APF), both DIOMs and DEOMs maintain their larval muscle morphology. (C) At 8 h APF the DEOMs, but not theDIOMS, begin to show signs of muscle degradation, as the muscles round up and the myosin pattern is disrupted. (D) By 12 h APF the DIOMs are more compact, but retain anorderly myosin pattern, while the DEOMs are almost completely disintegrated. (E–H) Electron micrographs (EM) of DIOMs from 0–12 h APF showing maintenance of thesarcomere throughout this period. (I–L) EM of DEOMs from 0–12 h APF. (I) At 0 h and (J) 4–6 h APF the muscles retain their larval sarcomere pattern, and vesicles occasionallyappear between the myofibrils (red arrow). (K) At 8 h APF, myofibrils are present, but are disorganized, while vesicles and multi-vesicular bodies (red arrows) are moreprevalent. (L) By 12 h APF, almost all myofibrils are gone. Abbreviations: Mf (myofibrils), Sr (sarcoplasmic reticulum), Mt (mitochondria).

J. Zirin et al. / Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 3

Please cite this article as: Zirin, J., et al., Ecdysone signaling at metamorphosis triggers apoptosis of Drosophila abdominal muscles. Dev.Biol. (2013), http://dx.doi.org/10.1016/j.ydbio.2013.08.029i

I-band (Fig. 1E and I). At 4–6 h APF both types of muscles retaintheir sarcomere structures, although sarcoplasmic reticulumand vesicles appear more frequently in the slightly expandedintermyofibrillar spaces of the DEOMs (Fig. 1F and J). At 8 h APFthe myofibrils of the DEOMs are still present, but are no longeraligned into well-ordered sarcomeres (Fig. 1K). The DIOMs at thesame stage maintain an orderly sarcomere structure (Fig. 1G).Finally, by 12 h APF, the DEOM myofibrils are mostly degraded(Fig. 1L), while the DIOM myofibrils are undegraded but abnor-mally aligned (Fig. 1H). Combined with the data obtained byconfocal microscopy, these EM pictures clearly show that themajor part of the DEOM histolysis occurs between 8 and 12 hAPF, during which time the muscles disintegrate and the myofi-brils are degraded.

Abdominal muscle histolysis is apoptotic, not autophagic

Previous studies of histolysis during Drosophila metamorphosishave shown that both apoptosis and autophagy play different rolesduring PCD, depending on the particular tissue. We thereforeexamined the roles of these two processes during DEOM histolysis(Fig. 2, Fig. S5). We first stained the abdominal muscles withα-human cleaved caspase-3, which recognizes multiple apoptoticproteins in Drosophila, and serves as a marker for the activity of theinitiator caspase Dronc (Fan and Bergmann, 2010). In control musclesexpressing a white dsRNA using the muscle specific Dmef2-Gal4driver line, neither the DIOMs nor the DEOMs show any caspase-3staining at 0 h APF (Fig. 2A). At 8 h APF however (Fig. 2B), consistentwith previous reports (Wasser et al., 2007), caspase-3 stains theDEOMs as they begin to histolyze, and the staining persists duringmuscle fragmentation (Fig. 2C) through their complete histolysis(Fig. 2D). During this period the DIOMs never stain positive forcaspase-3 (Fig. 2A–D). In addition, EM analysis of DEOMs at 8 h APFrevealed nuclei with condensed chromatin, one of the hallmarks ofapoptosis (Fig. 2M), although it is notable that the extent ofcondensation is not as severe as in other apoptotic cell types. Thismay be due to the unusual size of these nuclei, which have beengrowing via endoreplication throughout the larval stage. DNAfragmentation, identified by TUNEL, was also observed in DEOM,but not DIOM, nuclei at this time point (Fig. 2O) further implicatingthe apoptotic pathway during DEOM histolysis. Finally, we testedwhether blocking apoptosis would be sufficient to halt DEOMdegradation.Strikingly, overexpression of the caspase inhibitor p35 under controlof Dmef2-Gal4 abolished caspase-3 staining and strongly inhibitedDEOM histolysis and sarcomere degradation (Fig. 2E–H and P).

Studies in yeast have identified over 30 ATG (autophagy-related)genes, many of which are conserved in Drosophila and other higherorganisms (Chang and Neufeld, 2011; Yang and Klionsky, 2011;Yorimitsu and Klionsky, 2005; Zirin and Perrimon, 2010). In alleukaryotes, autophagy requires a functioning autophagy-related gene1 (Atg1) complex. We therefore knocked down Atg1 specifically inmuscles to determinewhether, like apoptosis, autophagy is involved inthe degradation of the DEOMs. In Dmef2-Gal4/UAS-Atg1 RNAi animalswe observed no delay in either caspase staining or muscle degrada-tion, which both began at 8 h APF, and continued as in control whiteRNAi muscles (Fig. 2I–L). We obtained the same results with RNAitargeting other essential autophagy components Atg5 and Atg18 (datanot shown). Additionally, each of these RNAi transgenes stronglyinhibited autophagosome formation in larval muscles, indicating thatthese reagents are effective at blocking autophagy (Fig. S2). We alsotested whether autophagosome formation was altered in the degrad-ing muscles. Double and single membrane vesicles containing cyto-plasm were observed in EM from DEOMs beginning at 6–8 h APF,suggesting increased autophagic/lysosomal activity (Fig. 2N, Fig. S6).Finally, localization of a GFP-tagged version of the conserved ubiquitin-

like protein Atg8 (Mizushima et al., 1998) was altered in the DEOMsrelative to DIOMs, although we were unable to discern the formationof GFP labeled autophagosomes in the histolyzing muscle (Fig. S2).Taken together, these results indicate that while autophagy may beincreased in DEOMs, it is not required for either caspase activation orthe degradation of this tissue.

Ecdysone signaling is required for abdominal muscle histolysis

Ecdysone signaling has been studied extensively in tissuesdestined for degradation or remodeling during Drosophila meta-morphosis (Bond et al., 2011; Cherbas et al., 2003; Jiang et al., 1997,2000; Lee and Baehrecke, 2001; Lee et al., 2003, 2002a, 2002b;Winbush and Weeks, 2011). However, the role of this pathwayduring abdominal muscle histolysis has not been tested. There arethree different isoforms of the EcR gene, EcR-A, EcR-B1, and EcR-B2, each sharing the same DNA-binding and ligand-bindingdomains, but with a unique amino terminus (Talbot et al., 1993).We first analyzed EcR expression during abdominal muscle degen-eration using antibodies specific to either EcR-A or EcR-B1 iso-forms. Both DEOMs and DIOMs express EcR-B1 throughout theperiod of muscle histolysis (Fig. 3A–D), while EcR-A was notexpressed in either muscle type (Fig. S3). We did not have anantibody that specifically recognizes EcR-B2, however, an antibodythat recognizes all EcR isoforms gave a staining pattern similar toα-EcR-B1 (Fig. S3), indicating that EcR-B2 is either absent from themuscles or has a pattern of expression similar to EcR-B1. Consis-tent with the expression data, knockdown of EcR-B1 using Dmef2-Gal4 caused a delay of DEOM degradation and caspase activation(Fig. 3E–H, Fig. S5). Further, knockdown with an RNAi reagenttargeting all EcR isoforms resulted in an even more substantialblock of degradation (Fig. 3I–J, Fig. S5), as did overexpression of adominant negative EcR-B1 transgene (Fig. 3K–L, Fig. S5). Giventhese results, we cannot rule out the possibility that EcR-B1 andEcR-B2 have some redundancy of function, which could accountfor the stronger phenotype in the dominant negative or all-isoform knockdown experiments.

FTZ-F1 is required for the proper timing of muscle histolysis

The stage specific expression of transcription factors is a criticaldeterminant of the response of cells to the ecdysone signal duringmetamorphosis. In particular, the orphan nuclear receptor FTZ-F1is subject to extensive regulation by the ecdysone signaling path-way during the prepupal stage, and is essential for the inductionsalivary gland histolysis (Thummel, 2001). We therefore examinedwhether FTZ-F1 similarly establishes the response of the abdom-inal muscle cells to the ecdysone signal. To follow the expressionof FTZ-F1, we used the protein trap line YB0007, in which GFP-tagged ftz-f1 is expressed under the control of its endogenouspromoter and enhancer elements (Buszczak et al., 2007). At 0 hAPF, FTZ-F1-GFP is not expressed in either the DIOMs or DEOMs(Fig. 4A). At 5 h APF we observed strong FTZ-F1-GFP expression inthe DEOMs, but none in the DIOMs (Fig. 4B). The fusion protein islocalized primarily to the nucleus, consistent with its role as atranscription factor. Expression of FTZ-F1-GFP is maintained in thedying DEOMs at 8 h APF, when it also appears in DIOMs, persistingin these muscles through 16 h APF (Fig. 4C–E). Thus, FTZ-F1expression is differentially regulated in the dying and persistentmuscles, and could therefore be an important determinant of theirdifferent responses to the ecdysone signal. To test this hypothesis,we knocked down ftz-f1 specifically in muscles using Dmef2-Gal4,and examined the death of DEOMs. Strikingly, we consistentlyobserved a delay in the activation of caspase and the death ofDEOMs in the ftz-f1 RNAi muscles (Fig. 4O–R, Fig. S5) compared towhite RNAi control muscles (Fig. 4. K–N, Fig. S5). In many cases, the

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132

J. Zirin et al. / Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎4

Please cite this article as: Zirin, J., et al., Ecdysone signaling at metamorphosis triggers apoptosis of Drosophila abdominal muscles. Dev.Biol. (2013), http://dx.doi.org/10.1016/j.ydbio.2013.08.029i

DEOMs persisted through 16 h APF with no detectable caspasestaining (Fig. 4R, Fig. S5). When we overexpressed FTZ-F1, theDEOMs degenerated prematurely, and the DIOMs had significantsarcomere disruption and reduced size (Fig. 4S–V, Fig. S5).However, FTZ-F1 overexpression was unable to induce earlycaspase activity in the DEOMs (Fig. 4S) or any caspase activity inthe DIOMs (Fig. 4S–V), indicating that FTZ-F1 expression alone isnot sufficient to trigger apoptosis of the abdominal muscles. It wasrecently shown that EcR-B1 expression depends on FTZ-F1 expres-sion in the neurons of the mushroom body during metamorphosis

(Boulanger et al., 2011). Given that we demonstrated that both ftz-f1 and EcR-B1 are essential for the proper histolysis of DEOMs, weexamined the expression of EcR-B1 in ftz-f1 RNAi muscles (Fig. S4).Unlike the mushroom body system, however, we could not detectchanges in EcR-B1 staining upon ftz-f1 knockdown.

HR39 opposes FTZ-F1 function during muscle histolysis

Like FTZ-F1, the nuclear receptor gene HR39 is expressed duringmetamorphosis in response to ecdysone (Horner et al., 1995). Both of

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132

Fig. 2. DEOM histolysis requires caspase activity but not autophagy. (A-L) Time course of abdominal muscles stained with Phalloidin/actin (green) and a-cleaved caspase3(red). In all panels the red outline indicates the DEOM and the blue outline the DIOM. (A-D) Muscles from control Dmef2-Gal4/UAS-white RNAi (UAS-white i) larvae. DIOMsshow no caspase activity at any time point, while the degrading DEOMs show caspase staining beginning at 8 h APF (B) through their disintegration at 12 h APF (C).(E–H) Muscles from Dmef2-Gal4/UAS-p35 larvae. DEOMs show no cleaved caspase staining and fail to degrade, although like the DIOMs, they become smaller by 16 h APF (H).(I–L) Muscles from Dmef2-Gal4/UAS-Atg1 RNAi larvae. Inhibition of autophagy has no effect on the timing of caspase activity in the DEOMs nor on their degradation. (M) EMof DEOM nucleus from a yw (control) larva at 8 h APF. Note the condensed chromatin characteristic of apoptosis. (N) EM of DEOM cytoplasm from a yw larva at 8 h APF. Thered arrow points to a vesicle containing cytoplasmic debris. (O) TUNEL staining of DIOM and DEOM from a yw larva at 8 h APF. TUNEL stain marks the shrunken nuclei of thedegrading DEOM (red arrow), but not the nuclei of the persistent DIOM (blue arrow). Blue¼DAPI, red¼TUNEL, green¼actin. (P) EM of DEOM from Dmef2-Gal4/UAS-p35larvae at 8 h APF. Expression of p35 inhibits the degradation of the sarcomere (compare to panel N). APF Abbreviations: Mf (myofibrils), Sr (sarcoplasmic reticulum),Mt (mitochondria), N (Nucleus).

J. Zirin et al. / Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 5

Please cite this article as: Zirin, J., et al., Ecdysone signaling at metamorphosis triggers apoptosis of Drosophila abdominal muscles. Dev.Biol. (2013), http://dx.doi.org/10.1016/j.ydbio.2013.08.029i

these factors have a high degree of sequence similarity, can bind to thesame DNA sequence, and regulate some of the same genes (Ayer et al.,1993; Ohno et al., 1994). Interestingly, Hr39 and ftz-f1 are expressed atroughly reciprocal timepoints in the prepupae, and have opposingphenotypes during mushroom body and motor neuronal remodeling(Boulanger et al., 2011, 2012; .Redt-Clouet et al., 2012; Sullivan andThummel, 2003). We hypothesized that, since FTZ-F1 plays animportant role in promoting abdominal muscle histolysis, HR39 mightplay the opposite role. We first examined the expression patter ofHr39 using a lacZ reporter (Fig. 4F–J). At 0 h APF, Hr39-lacZ isexpressed strongly in the DIOMs, and much more weakly in theDEOMs (Fig. 4F). At 5 h APF, when FTZ-F1 is present in the DEOMs, butabsent in the DIOMs, Hr39-lacZ has the opposite pattern, with onlyweak expression in the DEOMs and high expression in the DIOMs(Fig. 4G). From 8–16 h APF Hr39-lacZ expression continues in the

persistent DIOMs (Fig. 4H–J). Next, we overexpressed HR39, and foundthat it was able to repress DEOM histolysis and caspase activity(Fig. 4W–Z), much like in ftz-f1 knockdown muscles (Fig. 4O–R).Conversely, Hr39 mutant larvae prematurely activated caspase, has-tening the degradation of the DEOMs (Fig. 5A–E, Fig. S5). By 5 h APF(Fig. 5B), the DEOMs showed strong caspase activity, and themuscles were almost completely degraded by 8 h APF (Fig. 5C).However, reduced Hr39 had no apparent effect on the DIOMs(Fig. 5A–E). Finally, we examined the effect of muscle specific ftz-f1 knockdown in the Hr39 mutant animal to determine the epistaticrelationship of the two genes. With respect to DEOM degradation,we found no difference between ftz-f1 knockdown in the Hr39mutant trans-heterozygotes (Fig. 5K–O, Fig. S5) compared to ftz-f1knockdown in control Hr39/þ heterozygotes (Fig. 5F J, Fig. S5). Thatis, ftz-f1 knockdown inhibited caspase activity and DEOM histolysis,

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132

Fig. 3. Ecdysone signaling is required for DEOM histolysis. (A–D) Time course of abdominal muscles from a yw larvae stained with α-EcR-B1 (red), Phalloidin/actin (green), andDAPI (blue). In all panels the red outline indicates the DEOM and the blue outline indicates the DIOM. Both DEOMs and DIOMs express EcR-B1 throughout the period of musclehistolysis. (E–H) Abdominal muscles at 8 h APF and 16 h APF stained with Phalloidin/actin (green) and a-cleaved caspase3 (red). DEOMs from Dmef2-Gal4/UAS-EcR-B1 RNAi (G–H)show no cleaved caspase staining at 8 h APF and degradation is reduced at 16 h APF compared to Dmef2-Gal4 muscles (E–F). (I–J) Abdominal muscles at 16 h from Dmef2-Gal4/UAS-EcR RNAi104 larvae, stained with Phalloidin/actin (green). Both DIOM (I) and DEOM (J) persist with no signs of degradation. (K–L) Abdominal muscles at 16 h from Dmef2-Gal4/UAS-EcRDN larvae, stained with Phalloidin/actin (green). Similar to EcR knockdown, EcRDN expression causes both DIOM (K) and DEOM (L) to persist with no signs of degradation.

Fig. 4. FTZ-F1 and HR39 have opposing effects on DEOM histolysis. (A–E) Time course of abdominal muscles from Flytrap line YB0007, in which GFP-tagged FTZ-F1 isexpressed under the control of its endogenous promoter and enhancer elements. Larvae are stained with Phalloidin/actin (red). In all panels the blue outline indicates theDIOM and the yellow or red outline indicates the DEOM. At 5 h APF (B) FTZ-F1-GFP localizes to the nuclei of the DEOMs (yellow arrow) but is absent from the DIOMs (bluearrow). At 8 h APF (C) FTZ-F1-GFP localizes to both DIOMs and DEOMs, and persists in the DIOMs. (F–J) Time course of abdominal muscles expressing an Hr39-lacZ reporter,stained with Phalloidin/actin and DAPI (blue) and α-β-gal (red). At 0 h APF (F) and 8 h APF (G), Hr39-lacZ is expressed strongly in the DIOMs (blue arrows), but weakly in theDEOMs (red arrows). (H–J) HR39-lacZ expression persists in the DIOMs through 16 h APF. (K–Z) Time course of abdominal muscles stained with a-cleaved caspase3 (red) andPhalloidin/actin (green). (K–N) In control Dmef2-Gal4 animals, caspase staining appears at 8 h APF (L) in the degrading DEOMs. (O–R) In Dmef2-Gal4/UAS-ftz-f1 RNAi animals,caspase activation is repressed in DEOMs, which often persist through 16 h APF. (S–V) In Dmef2-Gal4/UAS-ftz-f1 over-expression animals, DEOMs are prematurely degradedat 8 h APF (T), while the DIOMs are thinner, but still fail to activate caspase through 16 h APF. (W–Z) In Dmef2-Gal4/UAS-Hr39 over-expression animals, neither DEOMs norDIOMs activate caspase or are degraded through 16 h APF.

J. Zirin et al. / Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎6

Please cite this article as: Zirin, J., et al., Ecdysone signaling at metamorphosis triggers apoptosis of Drosophila abdominal muscles. Dev.Biol. (2013), http://dx.doi.org/10.1016/j.ydbio.2013.08.029i

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132

J. Zirin et al. / Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 7

Please cite this article as: Zirin, J., et al., Ecdysone signaling at metamorphosis triggers apoptosis of Drosophila abdominal muscles. Dev.Biol. (2013), http://dx.doi.org/10.1016/j.ydbio.2013.08.029i

irrespective of the Hr39 background. Altogether, we conclude thatftz-f1 is epistatic to Hr39, consistent with the view that FTZ-F1expression is the ultimate determinant of DEOM death duringmetamorphosis.

Discussion

Ecdysone signaling is essential for histolysis of salivary glandand midgut, as well as for tissue remodeling during Drosophilametamorphosis (Bond et al., 2011; Cherbas et al., 2003; Jiang et al.,1997, 2000; Lee and Baehrecke, 2001; Lee et al., 2003, 2002a,2002b; Winbush and Weeks, 2011). However, the role of thispathway during abdominal muscle death had not previously beentested. There has been at least one report of ecdysone-independent PCD during metamorphosis (Hara et al., 2013) raisingthe possibility that muscle histolysis could be due to a differentsignaling pathway. In this study, we found that by knocking downall EcR isoforms specifically in the muscles, we could block muscledegradation, and that ectopic expression of a dominant negativeEcR-B1 isoform had the same effect, indicating that the ecdysonepathway functions cell-autonomously in the muscle to promotecell death during metamorphosis.

There are three different isoforms of the EcR gene, EcR-A, EcR-B1, and EcR-B2, each sharing the same DNA-binding and ligand-binding domains, but with a unique amino terminus (Talbot et al.,1993). The different temporal and spatial expression patterns ofEcR-A and EcR-B isoforms are thought to reflect their distinctfunctions during development (Davis et al., 2005; Truman et al.,1994). EcR-B1 is expressed primarily in larval cells that aredestined for histolysis, while EcR-A is expressed primarily inimaginal tissues destined for differentiation into adult structures(Talbot et al., 1993). Thus the response of salivary glands andmidgut to ecdysone during metamorphosis is dependent on EcR-

B1 (Bender et al., 1997). Nonetheless, EcR-A mutants also have adefect in salivary gland histolysis, suggesting that the isoformmight also contribute to this process (Davis et al., 2005). Further-more, some neurons in the ventral nerve cord and brain thatstrongly express EcR-A undergo apoptosis in response to ecdysonesoon after eclosion, suggesting that ecdysone induced PCD is notstrictly a function of EcR-B1 signaling (Kato et al., 2009; Robinowet al., 1993).

We examined expression of both EcR-A and EcR-B1 isoformsand found that only EcR-B1 was detectable in the DIOMs andDEOMs during pupariation. Consistent with its expression pattern,knockdown of EcR-B1 specifically in the muscle inhibited DEOMhistolysis. The inhibition achieved with the EcR-B1 isoform RNAiwas not as strong as with RNAi targeting all isoforms, or withoverexpression of the dominant negative EcR-B1. This could bedue to either differences in the efficiency of knockdown or to arole for EcR-B2 in DEOM degradation. Taken together our datastrongly supports the view that, like in salivary glands and midgut,ecdysone signals through EcR-B1 to induce cell death in abdominalmuscles. However, given that both DIOMs and DEOMs expressEcR-B1 at the same time, the presence of the receptor is notsufficient to explain why only the latter muscles are degraded.

Another important player in the timing of salivary glands celldeath is the orphan nuclear hormone receptor gene ftz-f1, which istranscribed midway through prepupal development, when theecdysone titer is relatively low (Lavorgna et al., 1993). FTZ-F1 hasbeen hypothesized as a competence factor, directing the subse-quent genetic responses to the ecdysone pulse at the prepupal/pupal transition (Broadus et al., 1999; Woodard et al., 1994). ThusFTZ-F1 is required for the induction of salivary gland histolysis. Incontrast, the midgut does not express ftz-f1 prior to its earlierecdysone induced cell death (Lee et al., 2002a), indicating thatFTZ-F1 is not required for all histolysis during Drosophila meta-morphosis. We found that despite the fact that the timing of

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132

Fig. 5. ftz-f1 is epistatic to Hr39 (A–O) Time course of abdominal muscles stained with a-cleaved caspase3 (red) and Phalloidin/actin (green). In all panels the blue outlineindicates the DIOM and the red outline the DEOM. (A–E) DEOMs from Hr393508/Hr39neo8; Dmef2-Gal4/TM6B larvae degenerate prematurely. At 5 h APF (B) the DIOMs haveactivated caspase, and by 8 h APF (C) the muscles are almost completely degraded. (F–J) DEOMs from Hr393508/CyO.GFP; Dmef2-Gal4/UAS-ftz-F1 RNAi larvae show delayedcaspase activation and degradation, similar to Dmef2-Gal4/UAS-ftz-F1 RNAi larvae (see Fig. 4O–R). (K–O) DEOMs from HR393508/HR39neo8; Dmef2-Gal4/UAS-ftz-F1 RNAi larvaeshow also show a delayed degradation phenotype, more like the ftz-f1 RNAi than the Hr39 mutant.

J. Zirin et al. / Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎8

Please cite this article as: Zirin, J., et al., Ecdysone signaling at metamorphosis triggers apoptosis of Drosophila abdominal muscles. Dev.Biol. (2013), http://dx.doi.org/10.1016/j.ydbio.2013.08.029i

muscle histolysis is similar to that of the midgut, the function ofFTZ-F1 was more like in the salivary gland, as FTZ-F1 was observedin the DEOMs starting at �5 h APF, prior to caspase activation at8 h APF. Furthermore, ftz-f1 was required for proper musclehistolysis, as knockdown cell-autonomously delayed caspase acti-vation and death in the DEOMs.

These results raised the possibility that the presence of FTZ-F1in the DEOMs, but not in the DIOMs, determined the differentresponse of these muscles to ecdysone. However, overexpressionof FTZ-F1 in the DIOMs, while causing severe muscle degeneration,was unable to induce caspase activity or cell death. Nor could thepresence of HR39 in the DIOMs account for the different responseof the muscles to ecdysone. Although we observed a reciprocity ofHR39 and FTZ-F1 expression in the DIOMs and DEOMs, Hr39mutant DIOMs still persist through metamorphosis. We concludethat FTZ-F1 and HR39 expression determine the timing of theabdominal muscle response to ecdysone but that these factors donot change the nature of the response.

It was recently shown that EcR-B1 expression is regulated by FTZ-F1 and HR39 in mushroom body neurons and abdominal motor-neurons during metamorphosis (Boulanger et al., 2011, 2012). Ourobservation that EcR-B1 promotes muscle degeneration is consistentwith the finding that EcR-B1 promotes post-synaptic dismantling inthe motor-neuron, and supports the notion that muscle degenerationis instructive on motor neuron retraction. However, we show thateven though both ftz-f1 and EcR-B1 are essential for the properhistolysis of DEOMs, there was no change in EcR-B1 staining uponftz-f1 knockdown as was observed in the mushroom body system.This suggests that changes observed in the muscle synapse due toftz-f1 knockdown are not the result of a downstream effect on EcR-B1expression in the muscle cell. Thus, the regulatory relationshipbetween FTZ-F1, HR39 and EcR-B1 in early pupal abdominal musclesis distinct from the relationship reported in neurons.

This suggests that there is an additional unknown factor whoseexpression dictates the fate of the DIOMs or DEOMs. This factor isunlikely to be either of the nuclear proteins EAST or Chromator(Chro), which were previously identified as having opposingeffects on the destruction of the abdominal DEOMs duringmetamorphosis (Wasser et al., 2007). Breakdown of DEOMs wasincomplete in Chro mutants, and promoted in east mutants,leading to the proposal that Chro activates and EAST inhibitstissue destruction and remodeling. However, neither east nor chroalleles cause histolysis of the DIOMs, nor do they alter caspaseactivation in either DEOMs or DIOMs. Rather these genes mayaffect the timing of muscle histolysis through a function down-stream of PCD induction. We propose that there must be anadditional factor present in the DIOMs which inhibits EcRB1signaling from inducing PCD, or alternatively, a factor present inthe DEOMs, which permits EcRB1 to activate a death program. Theidentification of this factor will be a focus of future studies.

The previously reported cleaved caspase-3 staining in theDEOMs is, to our knowledge, the only data addressing the natureof abdominal muscle PCD prior to our study. We aimed to addresswhether muscle histolysis is apoptotic, autophagic or some com-bination of both. During salivary gland histolysis, several autop-hagy related genes (ATGs) are upregulated (Gorski et al., 2003; Leeet al., 2003), and mutations or knockdown of these ATGs specifi-cally in the salivary gland inhibit the destruction of the tissue(Berry and Baehrecke, 2007). Caspase activation also occurs in thehistolyzing salivary glands, but overexpression of the caspaseinhibitor p35 only partially blocks salivary gland degradation(Lee and Baehrecke, 2001). Simultaneous inhibition of bothautophagy and apoptosis in the salivary gland produces thestrongest inhibition of death, suggesting that both pathwayscontribute to histolysis of this tissue (Berry and Baehrecke,2007). In contrast to salivary gland histolysis, midgut histolysis

requires autophagy but not caspase activity. Mutations or knock-down of ATGs inhibit midgut death, but p35 expression has noeffect (Denton et al., 2010, 2009). Based on these two modelsystems, it appears that there is no set rule as to the role ofautophagy and apoptosis during Drosophila histolysis.

Our data serves to further highlight how distinctive PCD for each ofthe tissues undergoing histolysis. We found that DEOMs stain positivefor cleaved caspase-3, consistent with previous reports. We alsoobserved TUNEL positive staining and chromatin condensation inthe DEOMs at 8 h APF, both markers of apoptosis. Importantly, wewere able to suppress DEOM degradation by overexpression of thepan-caspase inhibitor p35, indicating that unlike the midgut, musclehistolysis is apoptotic. To determine whether the muscle PCD wasautophagic in nature, we examined the DEOMs by EM. Although weobserved some autophagic vesicles in the dying muscles, they werenot abundant, nor were GFP-Atg8 localization to autophagosomesexamined by confocal microscopy. We knocked down several essentialcomponents of the autophagic machinery, and observed no effect onthe timing or extent of DEOM histolysis. Although knockdownefficiency is always a concern with RNAi experiments, each of thetransgenes was able to strongly inhibit autophagosome formation inlarval muscles. We are therefore confident that autophagy is notrequired for DEOM PCD, putting the abdominal muscle in the uniquecategory of non-autophagic histolysis. In future studies it will beinteresting to compare muscles, salivary gland, and midgut to deter-mine why each tissue has a distinctive type of PCD.

Acknowledgments

This work was supported by a RO1 grant (5R01AR057352) toNP; and an F32 NIH Postdoctoral Fellowship (5F32GM082174-02)to JZ. NP is an investigator of the HHMI. The work of JMD issupported by an ARC grant (SFI20121205950).

All of the transmission electron microscopy work was per-formed at the Electron Microscopy Unit, Department of ChemicalResearch Support of the Weizmann Institute of Science. We thankthe members of this unit for instruction and guidance. We are alsograteful to Prof. Ben-Zion Shilo and Dr. Eyal Schejter of theWeizmann Institute Department of Molecular Genetics for theiradvice and support.

Appendix A. Supplementary materials

Supplementary data associated with this article can be found inthe online version at http://dx.doi.org/10.1016/j.ydbio.2013.08.029.

References

Allen, A.K., Spradling, A.C., 2008. The Sf1-related nuclear hormone receptor Hr39regulates Drosophila female reproductive tract development and function.Development 135, 311–321.

Ayer, S., Walker, N., Mosammaparast, M., Nelson, J.P., Shilo, B.Z., Benyajati, C., 1993.Activation and repression of Drosophila alcohol dehydrogenase distal tran-scription by two steroid hormone receptor superfamily members binding to acommon response element. Nucleic Acids Res. 21, 1619–1627.

Baehrecke, E.H., 1996. Ecdysone signaling cascade and regulation of Drosophilametamorphosis. Arch. Insect Biochem. Physiol. 33, 231–244.

Baehrecke, E.H., 2005. Autophagy: dual roles in life and death? Nature reviews. Mol.Cell. Biol. 6, 505–510.

Baehrecke, E.H., Thummel, C.S., 1995. The Drosophila E93 gene from the 93F earlypuff displays stage- and tissue-specific regulation by 20-hydroxyecdysone. Dev.Biol. 171, 85–97.

Bate, M., Rushton, E., Currie, D.A., 1991. Cells with persistent twist expression arethe embryonic precursors of adult muscles in Drosophila. Development 113,79–89.

Bender, M., Imam, F.B., Talbot, W.S., Ganetzky, B., Hogness, D.S., 1997. Drosophilaecdysone receptor mutations reveal functional differences among receptorisoforms. Cell 91, 777–788.

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132

J. Zirin et al. / Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎ 9

Please cite this article as: Zirin, J., et al., Ecdysone signaling at metamorphosis triggers apoptosis of Drosophila abdominal muscles. Dev.Biol. (2013), http://dx.doi.org/10.1016/j.ydbio.2013.08.029i

Berry, D.L., Baehrecke, E.H., 2007. Growth arrest and autophagy are required forsalivary gland cell degradation in Drosophila. Cell 131, 1137–1148.

Bodenstein, D., TheQ4 postembryonic development of Drosophila.Bond, N.D., Nelliot, A., Bernardo, M.K., Ayerh, M.A., Gorski, K.A., Hoshizaki, D.K.,

Woodard, C.T., 2011. ssFTZ-F1 and Matrix metalloproteinase 2 are required forfat-body remodeling in Drosophila. Dev. Biol. 360, 286–296.

Boulanger, A., Clouet-Redt, C., Farge, M., Flandre, A., Guignard, T., Fernando, C.,Juge, F., Dura, J.M., 2011. ftz-f1 and Hr39 opposing roles on EcR expressionduring Drosophila mushroom body neuron remodeling. Nat. Neurosci. 14,37–44.

Boulanger, A., Farge, M., Ramanoudjame, C., Wharton, K., Dura, J.M., 2012. Drosophilamotor neuron retraction during metamorphosis is mediated by inputs from TGF-beta/BMP signaling and orphan nuclear receptors. PLoS One 7, e40255.

Broadus, J., McCabe, J.R., Endrizzi, B., Thummel, C.S., Woodard, C.T., 1999. TheDrosophila beta FTZ-F1 orphan nuclear receptor provides competence forstage-specific responses to the steroid hormone ecdysone. Mol. cell 3, 143–149.

Budnik, V., Gorczyca, M., Prokop, A., 2006. Selected methods for the anatomicalstudy of Drosophila embryonic and larval neuromuscular junctions. Int. Rev.Neurobiol. 75, 323–365.

Burtis, K.C., Thummel, C.S., Jones, C.W., Karim, F.D., Hogness, D.S., 1990. TheDrosophila 74EF early puff contains E74, a complex ecdysone-inducible genethat encodes two ets-related proteins. Cell 61, 85–99.

Buszczak, M., Paterno, S., Lighthouse, D., Bachman, J., Planck, J., Owen, S., Skora, A.D., Nystul, T.G., Ohlstein, B., Allen, A., Wilhelm, J.E., Murphy, T.D., Levis, R.W.,Matunis, E., Srivali, N., Hoskins, R.A., Spradling, A.C., 2007. The carnegie proteintrap library: a versatile tool for Drosophila developmental studies. Genetics 175,1505–1531.

Chang, Y.Y., Neufeld, T.P., 2011. Autophagy takes flight in Drosophila. FEBS Lett. 584,1342–1349.

Cherbas, L., Hu, X., Zhimulev, I., Belyaeva, E., Cherbas, P., 2003. EcR isoforms inDrosophila: testing tissue-specific requirements by targeted blockade andrescue. Development 130, 271–284.

Clarke, P.G., 1990. Developmental cell death: morphological diversity and multiplemechanisms. Anat. Embryol. 181, 195–213.

Clyne, P.J., Brotman, J.S., Sweeney, S.T., Davis, G., 2003. Green fluorescent proteintagging Drosophila proteins at their native genomic loci with small P elements.Genetics 165, 1433–1441.

Conradt, B., 2009. Genetic control of programmed cell death during animaldevelopment. Annu. Rev. Genet. 43, 493–523.

Currie, D.A., Bate, M., 1991. The development of adult abdominal muscles inDrosophila: myoblasts express twist and are associated with nerves. Develop-ment 113, 91–102.

Das, G., Shravage, B.V., Baehrecke, E.H., 2012. Regulation and function of autophagyduring cell survival and cell death. ColdQ5 Spring Harbor Perspectives Biol. 4.

Davis, M.B., Carney, G.E., Robertson, A.E., Bender, M., 2005. Phenotypic analysis ofEcR-A mutants suggests that EcR isoforms have unique functions duringDrosophila development. Dev. Biol. 282, 385–396.

Denton, D., Shravage, B., Simin, R., Baehrecke, E.H., Kumar, S., 2010. Larval midgutdestruction in Drosophila: not dependent on caspases but suppressed by theloss of autophagy. Autophagy 6, 163–165.

Denton, D., Shravage, B., Simin, R., Mills, K., Berry, D.L., Baehrecke, E.H., Kumar, S.,2009. Autophagy, not apoptosis, is essential for midgut cell death in Drosophila.Curr. Biol.: CB 19, 1741–1746.

DiBello, P.R., Withers, D.A., Bayer, C.A., Fristrom, J.W., Guild, G.M., 1991.The Drosophila Broad-Complex encodes a family of related proteins containingzinc fingers. Genetics 129, 385–397.

Dutta, D., Bloor, J.W., Ruiz-Gomez, M., VijayRaghavan, K., Kiehart, D.P., 2002. Real-time imaging of morphogenetic movements in Drosophila using Gal4-UAS-driven expression of GFP fused to the actin-binding domain of moesin. Genesis34, 146–151.

Fan, Y., Bergmann, A., 2010. The cleaved-Caspase-3 antibody is a marker of Caspase-9-like DRONC activity in Drosophila. Cell Death Differ. 17, 534–539.

Fernandes, J., Bate, M., Vijayraghavan, K., 1991. Development of the indirect flightmuscles of Drosophila. Development 113, 67–77.

Gorski, S.M., Chittaranjan, S., Pleasance, E.D., Freeman, J.D., Anderson, C.L., Varhol, R.J., Coughlin, S.M., Zuyderduyn, S.D., Jones, S.J., Marra, M.A., 2003. A SAGEapproach to discovery of genes involved in autophagic cell death. Curr. Biol.: CB13, 358–363.

Hara, Y., Hirai, K., Togane, Y., Akagawa, H., Iwabuchi, K., Tsujimura, H., 2013.Ecdysone-dependent and ecdysone-independent programmed cell death inthe developing optic lobe of Drosophila. Dev. Biol. 374, 127–141.

Horner, M.A., Chen, T., Thummel, C.S., 1995. Ecdysteroid regulation and DNAbinding properties of Drosophila nuclear hormone receptor superfamily mem-bers. Dev. Biol. 168, 490–502.

Jiang, C., Baehrecke, E.H., Thummel, C.S., 1997. Steroid regulated programmed celldeath during Drosophila metamorphosis. Development 124, 4673–4683.

Jiang, C., Lamblin, A.F., Steller, H., Thummel, C.S., 2000. A steroid-triggeredtranscriptional hierarchy controls salivary gland cell death during Drosophilametamorphosis. Mol. cell 5, 445–455.

Juhasz, G., Hill, J.H., Yan, Y., Sass, M., Baehrecke, E.H., Backer, J.M., Neufeld, T.P., 2008.The class III PI(3)K Vps34 promotes autophagy and endocytosis but not TORsignaling in Drosophila. J. Cell Biol. 181, 655–666.

Kato, K., Awasaki, T., Ito, K., 2009. Neuronal programmed cell death induces glialcell division in the adult Drosophila brain. Development 136, 51–59.

Kerr, J.F., Wyllie, A.H., Currie, A.R., 1972. Apoptosis: a basic biological phenomenonwith wide-ranging implications in tissue kinetics. Br. J. Cancer 26, 239–257.

Kimura, K.I., Truman, J.W., 1990. Postmetamorphic cell death in the nervous andmuscular systems of Drosophila melanogaster. J. Neurosci.: Off. J. Soc. forNeurosci. 10, 403. (401).

Koelle, M.R., Talbot, W.S., Segraves, W.A., Bender, M.T., Cherbas, P., Hogness, D.S.,1991. The Drosophila EcR gene encodes an ecdysone receptor, a newmember ofthe steroid receptor superfamily. Cell 67, 59–77.

Lavorgna, G., Karim, F.D., Thummel, C.S., Wu, C., 1993. Potential role for a FTZ-F1steroid receptor superfamily member in the control of Drosophila metamor-phosis. PNAS 90, 3004–3008.

Lee, C.Y., Baehrecke, E.H., 2001. Steroid regulation of autophagic programmed celldeath during development. Development 128, 1443–1455.

Lee, C.Y., Clough, E.A., Yellon, P., Teslovich, T.M., Stephan, D.A., Baehrecke, E.H., 2003.Genome-wide analyses of steroid- and radiation-triggered programmed celldeath in Drosophila. Curr. Biol.: CB 13, 350–357.

Lee, C.Y., Cooksey, B.A., Baehrecke, E.H., 2002a. Steroid regulation of midgut celldeath during Drosophila development. Dev. Biol. 250, 101–111.

Lee, C.Y., Simon, C.R., Woodard, C.T., Baehrecke, E.H., 2002b. Genetic mechanism forthe stage- and tissue-specific regulation of steroid triggered programmed celldeath in Drosophila. Dev. Biol. 252, 138–148.

Lockshin, R.A., Zakeri, Z., 2007. Cell death in health and disease. J.. Cell. Mol. Med.11, 1214–1224.

Mizushima, N., Noda, T., Yoshimori, T., Tanaka, Y., Ishii, T., George, M.D., Klionsky, D.J., Ohsumi, M., Ohsumi, Y., 1998. A protein conjugation system essential forautophagy. Nature 395, 395–398.

Ohno, C.K., Ueda, H., Petkovich, M., 1994. The Drosophila nuclear receptors FTZ-F1alpha and FTZ-F1 beta compete as monomers for binding to a site in the fushitarazu gene. Mol. Cell Biol. 14, 3166–3175.

Ranganayakulu, G., Schulz, R.A., Olson, E.N., 1996. Wingless signaling inducesnautilus expression in the ventral mesoderm of the Drosophila embryo. Dev.Biol. 176, 143–148.

Redt-Clouet, C., Trannoy, S., Boulanger, A., Tokmatcheva, E., Savvateeva-Popova, E.,Parmentier, M.L., Preat, T., Dura, J.M., 2012. Mushroom body neuronal remodel-ling is necessary for short-term but not for long-term courtship memory inDrosophila. Eur. J. Neurosci. 35, 1684–1691.

Riddiford, L.M., Cherbas, P., Truman, J.W., 2000. Ecdysone receptors and theirbiological actions. Vitamins and hormones 60, 1–73.

Robertson, C.W., 1936. The metamorphosis of Drosophila melanogaster, includingan accurately timed account of the principal morphological changes. J.Morphol. 59, 351–399.

Robinow, S., Talbot, W.S., Hogness, D.S., Truman, J.W., 1993. Programmed cell deathin the Drosophila CNS is ecdysone-regulated and coupled with a specificecdysone receptor isoform. Development 119, 1251–1259.

Roy, S., VijayRaghavan, K., 1998. Patterning muscles using organizers: larval muscletemplates and adult myoblasts actively interact to pattern the dorsal long-itudinal flight muscles of Drosophila. J. Cell Biol. 141, 1135–1145.

Segraves, W.A., Hogness, D.S., 1990. The E75 ecdysone-inducible gene responsiblefor the 75B early puff in Drosophila encodes two new members of the steroidreceptor superfamily. Genes Dev. 4, 204–219.

Sullivan, A.A., Thummel, C.S., 2003. Temporal profiles of nuclear receptor geneexpression reveal coordinate transcriptional responses during Drosophiladevelopment. Mol. Endocrinol. 17, 2125–2137.

Talbot, W.S., Swyryd, E.A., Hogness, D.S., 1993. Drosophila tissues with differentmetamorphic responses to ecdysone express different ecdysone receptor iso-forms. Cell 73, 1323–1337.

Thomas, H.E., Stunnenberg, H.G., Stewart, A.F., 1993. Heterodimerization of theDrosophila ecdysone receptor with retinoid X receptor and ultraspiracle.Nature 362, 471–475.

Thummel, C.S., 1995. From embryogenesis to metamorphosis: the regulation andfunction of Drosophila nuclear receptor superfamily members. Cell 83,871–877.

Thummel, C.S., 1996. Flies on steroids–Drosophila metamorphosis and the mechan-isms of steroid hormone action. Trends Genet.: TIG 12, 306–310.

Thummel, C.S., 2001. Molecular mechanisms of developmental timing in C. elegansand Drosophila. Dev. cell 1, 453–465.

Truman, J.W., Talbot, W.S., Fahrbach, S.E., Hogness, D.S., 1994. Ecdysone receptorexpression in the CNS correlates with stage-specific responses to ecdysteroidsduring Drosophila and Manduca development. Development 120, 219–234.

Wasser, M., Bte Osman, Z., Chia, W., 2007. EAST and Chromator control thedestruction and remodeling of muscles during Drosophila metamorphosis.Dev. Biol. 307, 380–393.

Winbush, A., Weeks, J.C., 2011. Steroid-triggered, cell-autonomous death of aDrosophila motoneuron during metamorphosis. Neural. Dev. 6, 15.

Woodard, C.T., Baehrecke, E.H., Thummel, C.S., 1994. A molecular mechanism forthe stage specificity of the Drosophila prepupal genetic response to ecdysone.Cell 79, 607–615.

Yang, Z., Klionsky, D.J., 2009. An overview of the molecular mechanism ofautophagy. Curr. Top. Microbiol. Immunol. 335, 1–32.

Yang, Z., Klionsky, D.J., 2011. Mammalian autophagy: core molecular machinery andsignaling regulation. Curr. Opin. Cell Biol. 22, 124–131.

Yao, T.P., Segraves, W.A., Oro, A.E., McKeown, M., Evans, R.M., 1992. Drosophilaultraspiracle modulates ecdysone receptor function via heterodimer formation.Cell 71, 63–72.

Yorimitsu, T., Klionsky, D.J., 2005. Autophagy: molecular machinery for self-eating.Cell Death Differ. 12 (Suppl. 2), 1542–1552.

Zirin, J., Perrimon, N., 2010. Drosophila as a model system to study autophagy.Semin. Immunopathol. 32, 363–372.

123456789

101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566

676869707172737475767778798081828384858687888990919293949596979899

100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132

J. Zirin et al. / Developmental Biology ∎ (∎∎∎∎) ∎∎∎–∎∎∎10

Please cite this article as: Zirin, J., et al., Ecdysone signaling at metamorphosis triggers apoptosis of Drosophila abdominal muscles. Dev.Biol. (2013), http://dx.doi.org/10.1016/j.ydbio.2013.08.029i